16 messages in org.freebsd.freebsd-currentRe: suspend bug | From | Sent On | Attachments |
|---|---|---|
| Bakul Shah | Jul 18, 2002 12:33 pm | |
| Bruce Evans | Jul 20, 2002 9:32 pm | |
| Bruce Evans | Jul 20, 2002 9:53 pm | |
| Bakul Shah | Jul 20, 2002 11:17 pm | |
| Julian Elischer | Jul 20, 2002 11:35 pm | |
| Julian Elischer | Jul 20, 2002 11:40 pm | |
| Andrey A. Chernov | Jul 21, 2002 1:32 am | |
| Bruce Evans | Jul 21, 2002 3:29 am | |
| Tim Robbins | Jul 21, 2002 4:09 am | |
| David Xu | Jul 21, 2002 4:21 am | |
| Andrey A. Chernov | Jul 21, 2002 6:32 am | |
| Andrey A. Chernov | Jul 21, 2002 5:24 pm | |
| Tim Robbins | Jul 21, 2002 5:31 pm | |
| Julian Elischer | Jul 21, 2002 10:37 pm | |
| Bruce Evans | Jul 22, 2002 7:02 am | |
| Julian Elischer | Jul 23, 2002 12:18 pm |
| Subject: | Re: suspend bug | |
|---|---|---|
| From: | Bakul Shah (bak...@bitblocks.com) | |
| Date: | Jul 20, 2002 11:17:19 pm | |
| List: | org.freebsd.freebsd-current | |
Some more details: - this happens for csh but not for bash-1. - ktracing everything shows further strange behaviour. ktrace causes a kernel printf "Out of ktrace request objects.". The login shell is not lost immediately. You get a shell prompt, but at least tty echo is broken, and hitting return a few times causes all the shells to exit.
With zsh the login shell doesn't die but the suspended su stays suspended after repeated fg attempts.
The bug is actually in su. It happens with a pre-KSEIII kernel. Backing out rev.1.54 of su.c (which was apparently supposed to fix a related bug tcsh's fg command) fixes it.
Thanks very much for tracking this down! Should've suspected su first (sorry, Julian!).
It would be nice if we had a regression test for most commands to prevent such breakage. If there is a framework in place people can add tests for individual commands. Ideas, anyone?
-- bakul
To Unsubscribe: send mail to majo...@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message